Research from Asahi University Hospital Has Provided New Data on Artificial Intelligence (Non-polypoid Colorectal Lesions Detection and False Positive Detection by Artificial Intelligence under Blue Laser Imaging and Linked Color Imaging).

A recent study conducted at Asahi University Hospital has found that artificial intelligence (AI) with white light imaging (WLI) is not sufficient for detecting non-polypoid colorectal polyps and has a high false positive rate (FPR). The researchers developed AIs using blue laser imaging (BLI) and linked color imaging (LCI) with specific learning sets (LS) to improve detection. The study compared the sensitivity and FPR of different AI models and found that AI finetuned by appropriate LS was able to detect non-reddish and non-polypoid polyps under LCI. This research provides valuable insights into the potential of AI in colorectal research and healthcare. [Extracted from the article]